Prostate cancer volume correlates with stage, grade, and progression after prostatectomy. When tumor volume is measured planimetrically, results are multiplied by a correction factor to compensate for tissue shrinkage caused by processing. Injection of formalin into prostatectomy specimens was suggested for improved fixation. Our aim was to investigate how this affects the prostate volume. We studied 142 radical prostatectomy specimens. All prostates were immersed in 10% formalin. In 84 prostates (59%) we also injected 20 ml of formalin before routine fixation. The prostates were weighed unfixed after injection and after final fixation. The specimens were sliced and totally embedded. The transverse diameters of the prostates were measured on unfixed specimens and microscopic sections. The average weight loss after final fixation was 5.8 and 8.6% for formalin-injected specimens and standard-fixed specimens, respectively (p<0.001). However, when total shrinkage was estimated from the transverse diameters, there was no difference related to fixation technique (p=0.59). The average linear shrinkage was 4.5%, corresponding to a volume correction factor of 1.15. We conclude that formalin injection for fixation of prostate tissue does not influence tumor volume calculation compared to conventional fixation.

Heat shock proteins (HSPs) protect cells against stress-associated injury and are overexpressed in several malignant tumors. We aimed to investigate their value as prognostic markers in prostate cancer. A tissue microarray (TMA) was constructed of 289 prostate cancers from radical prostatectomy (RP) specimens with median follow-up of 48.9 months. Slides were immunostained for HSP27, HSP60 and HSP70. Intensity and extent of immunoreactivity (IR) and their product (IRp) was evaluated by two observers. The IRp of HSP27 and HSP60, but not of HSP70, significantly predicted biochemical recurrence (p=0.014, 0.034 and 0.160, respectively). Recurrence-free survival in patients with strong HSP27 and HSP60 staining was shorter than in those with weak expression (p=0.019 and 0.001, respectively). IRp of HSP27 and HSP60 correlated with Gleason score (p<0.01). HSP60 was an independent predictor of biochemical recurrence in multivariate analysis, including extraprostatic extension, margin status, seminal vesicle invasion and Gleason score. Weighted kappa for interobserver agreement of HSP27, HSP60 and HSP70 IR was 0.613-0.823 for intensity and 0.584-0.719 for IRp, but only 0.036-0.244 for extent, raising the question whether staining extent should be estimated on TMA. We conclude that HSP27 and HSP60 are predictors of biochemical recurrence after RP.

Pancreatic duodenal homeobox 1 (PDX-1), a Hox type transcription factor, is necessary for differentiation of exocrine and endocrine pancreas, and regulates insulin gene transcription. PDX-1 expression was studied by immunohistochemistry on a tissue microarray (TMA) of 289 primary prostate cancers (PCa) from radical prostatectomy (RP) specimens with median follow-up of 48.9 months. We separately arrayed benign prostatic tissue, atrophy, high-grade prostatic intraepithelial neoplasia (HGPIN) and PCa from 40 men and also 17 lymph node metastases. Intensity and extent of immunoreactivity and their product (IRp) were evaluated by two independent observers. PDX-1 was overexpressed in cancer vs benign tissue (p<0.001), but also in atrophy and HGPIN vs cancer (p<0.001 and p=0.022, respectively). PDX-1 expression did not correlate with biochemical recurrence, but decreased with higher Gleason pattern (p<0.001) and in metastases vs primary PCa (p<0.001). Weighted kappa for interobserver agreement of intensity, extent and IRp was 0.65, 0.13 and 0.54, respectively. Presence of PDX-1 protein in benign and malignant prostatic tissue was confirmed by Western blot. In view of recent attention to the role of insulin systems in men with PCa, this protein is of interest in the pathogenesis of PCa.

The management and prognostication of patients with urothelial carcinomas (UCs), the most common histological type of bladder cancer, is mainly based on clinicopathological parameters. Several markers have been proposed to monitor this disease, including individual cell cycle-related proteins such as p53, pRb, p16, p21 and p27. Other putative markers are the oncogene products of FGFR3 and the ErbB family, proliferation markers including Ki-67, Aurora-A and survivin and different components within the immune system. In this review, a total of 12 parameters were evaluated and their discriminatory power compared. It is concluded that, in single-marker analyses, the proliferation markers Ki-67, survivin and Aurora-A offer the best potential to predict disease progression since they were all able to demonstrate independent prognostic power in repeated studies. Markers related to the immune system (e.g. CD8+ cells, regulatory T cells and cyclooxygenase-2 expression) or oncogene products of the ErbB family and FGFR3 are less powerful predictors of outcome or have not been equally well studied. The cell cycle-related proteins p53, pRb, p16, p21 and p27 have been extensively studied, but their usefulness as single prognostic markers remains unclear. However, in multimarker analyses, these markers appear to add prognostic information, indicating that they may contribute to more accurate treatment of UC.

No abstract